1
|
AlFaraj Y, Mohapatra S, Shieh P, Husted KEL, Ivanoff DG, Lloyd EM, Cooper JC, Dai Y, Singhal AP, Moore JS, Sottos NR, Gomez-Bombarelli R, Johnson JA. A Model Ensemble Approach Enables Data-Driven Property Prediction for Chemically Deconstructable Thermosets in the Low-Data Regime. ACS CENTRAL SCIENCE 2023; 9:1810-1819. [PMID: 37780353 PMCID: PMC10540282 DOI: 10.1021/acscentsci.3c00502] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 04/20/2023] [Indexed: 10/03/2023]
Abstract
Thermosets present sustainability challenges that could potentially be addressed through the design of deconstructable variants with tunable properties; however, the combinatorial space of possible thermoset molecular building blocks (e.g., monomers, cross-linkers, and additives) and manufacturing conditions is vast, and predictive knowledge for how combinations of these molecular components translate to bulk thermoset properties is lacking. Data science could overcome these problems, but computational methods are difficult to apply to multicomponent, amorphous, statistical copolymer materials for which little data exist. Here, leveraging a data set with 101 examples, we introduce a closed-loop experimental, machine learning (ML), and virtual screening strategy to enable predictions of the glass transition temperature (Tg) of polydicyclopentadiene (pDCPD) thermosets containing cleavable bifunctional silyl ether (BSE) comonomers and/or cross-linkers with varied compositions and loadings. Molecular features and formulation variables are used as model inputs, and uncertainty is quantified through model ensembling, which together with heavy regularization helps to avoid overfitting and ultimately achieves predictions within <15 °C for thermosets with compositionally diverse BSEs. This work offers a path to predicting the properties of thermosets based on their molecular building blocks, which may accelerate the discovery of promising plastics, rubbers, and composites with improved functionality and controlled deconstructability.
Collapse
Affiliation(s)
- Yasmeen
S. AlFaraj
- Department
of Chemistry, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, United States of America
| | - Somesh Mohapatra
- Department
of Materials Science and Engineering, Massachusetts
Institute of Technology, Cambridge, Massachusetts 02139, United States of America
| | - Peyton Shieh
- Department
of Chemistry, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, United States of America
| | - Keith E. L. Husted
- Department
of Chemistry, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, United States of America
| | - Douglass G. Ivanoff
- Department
of Materials Science and Engineering, University
of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States of America
- The
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States
of America
| | - Evan M. Lloyd
- The
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States
of America
- Department
of Chemistry, University of Illinois at
Urbana—Champaign, Urbana, Illinois 61801, United States of America
| | - Julian C. Cooper
- The
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States
of America
- Department
of Chemistry, University of Illinois at
Urbana—Champaign, Urbana, Illinois 61801, United States of America
| | - Yutong Dai
- Department
of Chemistry, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, United States of America
| | - Avni P. Singhal
- Department
of Materials Science and Engineering, Massachusetts
Institute of Technology, Cambridge, Massachusetts 02139, United States of America
| | - Jeffrey S. Moore
- Department
of Materials Science and Engineering, University
of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States of America
- The
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States
of America
| | - Nancy R. Sottos
- Department
of Materials Science and Engineering, University
of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States of America
- The
Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana—Champaign, Urbana, Illinois 61801, United States
of America
| | - Rafael Gomez-Bombarelli
- Department
of Materials Science and Engineering, Massachusetts
Institute of Technology, Cambridge, Massachusetts 02139, United States of America
| | - Jeremiah A. Johnson
- Department
of Chemistry, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139, United States of America
| |
Collapse
|
2
|
Fan Z, Wang Y, Ying P, Song K, Wang J, Wang Y, Zeng Z, Xu K, Lindgren E, Rahm JM, Gabourie AJ, Liu J, Dong H, Wu J, Chen Y, Zhong Z, Sun J, Erhart P, Su Y, Ala-Nissila T. GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations. J Chem Phys 2022; 157:114801. [DOI: 10.1063/5.0106617] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/14/2022] Open
Abstract
We present our latest advancements of machine-learned potentials (MLPs) based on the neuroevolution potential (NEP) framework introduced in [Fan et al., Phys. Rev. B 104, 104309 (2021)] and their implementation in the open-source package GPUMD.We increase the accuracy of NEP models both by improving the radial functions in the atomic-environment descriptor using a linear combination of Chebyshev basis functions and by extending the angular descriptor with some four-body and five-body contributions as in the atomic cluster expansion approach.We also detail our efficient implementation of the NEP approach in graphics processing units as well as our workflow for the construction of NEP models, and we demonstrate their application in large-scale atomistic simulations.By comparing to state-of-the-art MLPs, we show that the NEP approach not only achieves above-average accuracy but also is far more computationally efficient.These results demonstrate that the GPUMD package is a promising tool for solving challenging problems requiring highly accurate, large-scale atomistic simulations.To enable the construction of MLPs using a minimal training set, we propose an active-learning scheme based on the latent space of a pre-trained NEP model.Finally, we introduce three separate Python packages, GPYUMD, CALORINE, and PYNEP, which enable the integration of GPUMD into Python workflows.
Collapse
Affiliation(s)
- Zheyong Fan
- School of Mathematics and Physics, Bohai University, China
| | | | - Penghua Ying
- School of Science, Harbin Institute of Technology Shenzhen, China
| | - Keke Song
- University of Science and Technology Beijing, China
| | | | | | | | - Ke Xu
- Xiamen University, Xiamen University, China
| | | | | | | | - Jiahui Liu
- University of Science and Technology Beijing, China
| | | | - Jianyang Wu
- Department of Physics, Xiamen University, China
| | - Yue Chen
- Mechanical Engineering, University of Hong Kong Department of Mechanical Engineering, Hong Kong
| | - Zheng Zhong
- Harbin Institute of Technology, Shenzhen, Harbin Institute of Technology, China
| | - Jian Sun
- Department of Physics and National Laboratory of Solid State Microstructures, Nanjing University, China
| | | | - Yanjing Su
- Corrosion and Protection Center, Key Laboratory for Environmental Fracture (MOE), University of Science and Technology Beijing, China
| | - Tapio Ala-Nissila
- Department of Applied Physics, Aalto University Department of Applied Physics, Finland
| |
Collapse
|
3
|
Zaverkin V, Holzmüller D, Schuldt R, Kästner J. Predicting properties of periodic systems from cluster data: A case study of liquid water. J Chem Phys 2022; 156:114103. [DOI: 10.1063/5.0078983] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/14/2022] Open
Abstract
The accuracy of the training data limits the accuracy of bulk properties from machine-learned potentials. For example, hybrid functionals or wave-function-based quantum chemical methods are readily available for cluster data but effectively out of scope for periodic structures. We show that local, atom-centered descriptors for machine-learned potentials enable the prediction of bulk properties from cluster model training data, agreeing reasonably well with predictions from bulk training data. We demonstrate such transferability by studying structural and dynamical properties of bulk liquid water with density functional theory and have found an excellent agreement with experimental and theoretical counterparts.
Collapse
Affiliation(s)
- Viktor Zaverkin
- Institute for Theoretical Chemistry, University of Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany
| | - David Holzmüller
- Institute for Stochastics and Applications, University of Stuttgart, Pfaffenwaldring 57, 70569 Stuttgart, Germany
| | - Robin Schuldt
- Institute for Theoretical Chemistry, University of Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany
| | - Johannes Kästner
- Institute for Theoretical Chemistry, University of Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany
| |
Collapse
|
4
|
Zaverkin V, Holzmüller D, Steinwart I, Kästner J. Fast and Sample-Efficient Interatomic Neural Network Potentials for Molecules and Materials Based on Gaussian Moments. J Chem Theory Comput 2021; 17:6658-6670. [PMID: 34585927 DOI: 10.1021/acs.jctc.1c00527] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022]
Abstract
Artificial neural networks (NNs) are one of the most frequently used machine learning approaches to construct interatomic potentials and enable efficient large-scale atomistic simulations with almost ab initio accuracy. However, the simultaneous training of NNs on energies and forces, which are a prerequisite for, e.g., molecular dynamics simulations, can be demanding. In this work, we present an improved NN architecture based on the previous GM-NN model [Zaverkin V.; Kästner, J. J. Chem. Theory Comput. 2020, 16, 5410-5421], which shows an improved prediction accuracy and considerably reduced training times. Moreover, we extend the applicability of Gaussian moment-based interatomic potentials to periodic systems and demonstrate the overall excellent transferability and robustness of the respective models. The fast training by the improved methodology is a prerequisite for training-heavy workflows such as active learning or learning-on-the-fly.
Collapse
Affiliation(s)
- Viktor Zaverkin
- Institute for Theoretical Chemistry, University of Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany
| | - David Holzmüller
- Institute for Stochastics and Applications, University of Stuttgart, Pfaffenwaldring 57, 70569 Stuttgart, Germany
| | - Ingo Steinwart
- Institute for Stochastics and Applications, University of Stuttgart, Pfaffenwaldring 57, 70569 Stuttgart, Germany
| | - Johannes Kästner
- Institute for Theoretical Chemistry, University of Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany
| |
Collapse
|